Bovine leukemia virus (BLV) is an oncogenic retrovirus that causes lymphoid cancer in cattle with a long latency after infection. BLV and the human T- cell leukemia viruses (HTLV) types I and II form a distinct group of oncogenic retroviruses whose mechanism of tumorigenesis is unknown. Their genomes are structurally homologous; each encodes at least two nonvirion proteins that regulate viral replication. Transcriptional activation of host cell genes by one of these proteins has been proposed as a factor contributing to increased host cell proliferation. The viruses share many features of infection and pathogenesis, although BLV causes B-cell tumors whereas HTLV causes T-cell tumors. These viruses establish a latent but persistent infection in their hosts. The identification of factors that activate viral expression from latency is central to understanding pathogenesis. The similarities between the viruses warrant study of BLV because it can be experimentally manipulated in vivo. The goal of this research is to understand how BLV gene expression and virus production are regulated in the cells that host latent infections in vivo. The stimuli encountered by host lymphocytes when participating in immune responses may be important regulators of viral gene expression. Peripheral blood mononuclear cells obtained from sheep in known stages of BLV infection will be manipulated in culture to test the hypothesis that the state of cellular activation differentially supports BLV expression at transcription and translation of viral messenger RNAs, and at virus production. To determine how activation of B cells by the antigen- lymphokine pathway supports BLV replication, cells will be stimulated with mitogens or anti-immunoglobulins and cultured in T cell-conditioned medium. Virus expression will be assessed by in situ hybridization for viral RNA, by immunocytochemistry for viral proteins, and by infectious centers assays for virus production. Many BLV-infected cells express the viral caps protein after short-term culture, but some fail to synthesize the virion surface glycoprotein required for particle infectivity. Experiments will be performed to determine whether messenger RNA availability explains this difference and whether T cell-conditioned medium can selectively enhance surface glycoprotein synthesis. The polymerase chain reaction will be used to discriminate the three classes of viral mRNA in populations in which infected cells are scarce. As they come out of the circulation, BLV- infected cells exist in several states with respect to the stimuli required to induce viral expression. Since a proportion of circulating sheep B cells are the progency of recent mitosis, the ability of these cells to preferentially support BLV replication will be assessed.